Investment banks tend to underprice IPOs by about 10-15% on average, and intentionally so. This makes allocation a favor, because of the obvious expectancy in being in on one. However, the company ends up being undercapitalized for the amount of equity given up. It's utterly sleazy, but it's business as usual on Wall Street.

This also means that if you're not on a bank's favorite clients list, you don't want to be involved with IPOs, because you're only going to be able to get in on the crappy ones.

The concept of shorting Facebook is somewhat of a joke, because of that. I'm no fan of Mark Zuckerberg, but I wouldn't actually claim to know better than the market what the company's worth. That said, if there were a way to have shorted some share of the company at $15B when Microsoft bought in, that would have been awesome.